Stable solid-supported leucoanthocyanidin variants for flavanoid biosynthesis elucidation

Article abstract of DOI:10.1016/j.tetlet.2009.09.045

Remarkable progress toward the complete elucidation of the biosynthesis of flavanoids has been accomplished during the last decade, but the final steps presumably involving the transformation of leucoanthocyanidins, which are highly unstable when free in aqueous solution, into both anthocyanidins and proanthocyanidins still remain to be fully understood. Herein is described the synthesis of stabilized solid-supported leucoanthocyanidin variants that should serve as valuable tools for in vitro studies aimed at investigating the metabolism of these seemingly fleeting common precursors of two main classes of flavanoids.

Full text of DOI:10.1016/j.tetlet.2009.09.045

Tetrahedron Letters 50 (2009) 6567–6571
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Stable solid-supported leucoanthocyanidin variants for flavanoid
biosynthesis elucidation
Denis Deffieux a,b, , Sophie Gaudrel-Grosay a,b, Axelle Grelard b, Céline Chalumeau a,b, Stéphane Quideau a,b,
*
*
a Université de Bordeaux, Institut des Sciences Moléculaires (CNRS-UMR 5255), 351 cours de la libération, 33405 Talence Cedex, France
b Institut Européen de Chimie et Biologie, 2 rue Robert Escarpit, 33607 Pessac Cedex, France
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 20 July 2009
Revised 7 September 2009
Accepted 9 September 2009
Available online 12 September 2009
Remarkable progress toward the complete elucidation of the biosynthesis of flavanoids has been accom-
plished during the last decade, but the final steps presumably involving the transformation of leucoanth-
ocyanidins, which are highly unstable when free in aqueous solution, into both anthocyanidins and
proanthocyanidins still remain to be fully understood. Herein is described the synthesis of stabilized
solid-supported leucoanthocyanidin variants that should serve as valuable tools for in vitro studies aimed
at investigating the metabolism of these seemingly fleeting common precursors of two main classes of
flavanoids.
Keywords:
Flavanoids
Leucoanthocyanidins
Anthocyanidins
Proanthocyanidins
Biosynthesis
Ó 2009 Elsevier Ltd. All rights reserved.
The biosynthesis of flavo/flavanoids is probably one of the most
extensively studied plant metabolic pathways.1 Most of the genes
encoding the enzymes that are responsible for the catalysis of each
individual step have been isolated from different plant sources.2
However, the last steps leading to the formation of flavylium-based
anthocyani(di)ns and flavan-3-ol-derived proanthocyanidins re-
main unclear and are still under debate due in part to the lack of
in vitro evidence.3 The putative biosynthesis pathway leading to
these two important classes of flavanoids4 is depicted in Scheme 1.
Leucoanthocyanidins [i.e., (2R,3S,4S)-flavan-2,3-trans-3,4-cis-diols]
arepresumedto be thecommonprecursorsof bothanthocyani(di)ns
and proanthocyanidins. Results from in vitro studies have shown
that Leucoanthocyanidin DiOXygenase (LDOX, also referred to as
ANS for ANthocyanidin Synthase), a 2-oxoglutarate-dependent en-
zyme, is involved in the conversion of colorless leucoanthocyanidins
into colored anthocyanidins 3-O-glucosides.5 These anthocyanins
can then undergo methylation of their ring-B hydroxyl groups.
However, the mechanistic details of the conversion of leucoanth-
ocyanidins into anthocyanidins have not yet been fully elucidated
due to the lack of structural information on the intermediate(s)
entailed in this biotransformation.6 Even more speculative are the
final stages of the biosynthesis of proanthocyanidins.7 These so-
called condensed tannins would result from the polymerization of
flavan-3-ols and/or flavan-3,4-diols (Scheme 1). At least two addi-
tional enzymes would be involved in the process leading to flavan-
3-ols from leucoanthocyanidins. Thus, (2R,3R)-cis-flavan-3-ols such
as (ꢀ)-epicatechin indirectly derive from leucoanthocyanidins
through the action of an ANthocyanidin Reductase (ANR) on antho-
cyanidins,3a whereas a LeucoAnthocyanidin Reductase (LAR) di-
rectly converts leucoanthocyanidins into (2R,3S)-trans-flavan-3-ols
such as (+)-catechin.8
Several questions about the (bio)polymerization leading to pro-
anthocyanidins still remain unanswered such as those concerning
the exact nature of the starter and extension units, as well as
whether or not a dedicated enzyme (i.e., a ‘proanthocyanidin syn-
thase’) is involved in the process.7,9 The keystone position of leuco-
anthocyanidins in this biosynthetic scheme obviously makes them
substrates of choice for in vitro experiments aimed at addressing
all of the above issues remaining obscure in these final steps of
the biogenesis of anthocyani(di)ns and proanthocyanidins. Unfor-
tunately, such experiments are far from being easy to conduct be-
cause of the notable instability of these flavan-3,4-cis-diol
compounds, which are prone to polymerize even under slightly
acidic conditions.1a,10 In aqueous solutions, highly reactive (pro-
tonated) para- and/or ortho-quinone methides resulting from a fac-
ile departure of the benzylic hydroxyl group at position 4 are
probably responsible for the rapid formation of polymeric materi-
als in an uncontrolled manner. This is also probably the main rea-
son why 3,4-cis leucoanthocyanidins that have been previously
generated by chemical synthesis were characterized under their
protected methyl ether forms.11 The very few enzymatic studies
that have been carried out to date with these compounds in fact
report the use of unprotected (2R,3S,4R)-diastereomers of
* Corresponding authors. Tel.: +33 540 00 30 10; fax: +33 540 00 22 15 (S.Q.).
E-mail address: s.quideau@iecb.u-bordeaux.fr (S. Quideau).
0040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.09.045

6568
D. Deffieux et al. / Tetrahedron Letters 50 (2009) 6567–6571
thus immobilized at an appropriate mutual distance on an insoluble
R1
R1
cross-linked polymer, the selected leucoanthocyanidin variants
should find themselves in conditions approaching those of high (infi-
nite) dilution, hence minimizing the possibilities of intermolecular
reactions.13
OH
R2
OH
R2
B
2
HO
O
HO
O
LDOX
A
C
3
OR3
OH
4
A synthesis of solid-supported leucoanthocyanidins relying on
an early grafting step using the polymer support as a protecting
group could have constituted a more elegant strategy, but the dif-
ficulty of structurally characterizing polymer-bound intermediates
by means of spectroscopical methods led us to favor a more prac-
tical approach based on the synthesis of the desired leucoanthocy-
anidin molecules prior to their loading onto the polymer support.
The synthesis of the racemic protected leucoanthocyanidins 7a–c
was thus first accomplished (Scheme 2). In each of the three cases,
the phenolic hydroxyl function at the 40-position of ring-B had to
be orthogonally protected in anticipation of its exploitation for
the attachment to the polymer support. In addition, the protecting
groups for all of the other hydroxyl functions had to be chosen
while keeping in mind the necessity of removing them under mild
conditions without perturbing the fragile leucoanthocyanidin mol-
ecules still attached to the polymer support. To meet these require-
ments, silyl groups were selected to protect the hydroxyl functions
at positions 3, 4, 5, and 7, whereas a photolabile 6-nitroveratryl or,
alternatively, a classic benzyl group was used for the orthogonal
protection of the phenolic 40-hydroxyl function.
The syntheses started with the partial silylation of commercially
available 2,4,6-trihydroxyacetophenone using triisopropylsilyl
triflate (TIPSOTf) and triethylamine (Et3N) in CH2Cl2 at low tempera-
ture to furnish the phenolic ketone 1 in high yield. The phenolic
hydroxyl group of commercial syringaldehyde was protected via a
Williamson reaction using 4,5-dimethoxy-2-nitrobenzyl bromide to
furnish 2a in 97% yield (see Supplementary data). The other two
ring-B precursors, that is, 4-benzyloxy-3-methoxybenzaldehyde
(2b)and4-benzyloxybenzaldehyde(2c),arebothcommerciallyavail-
able. The phenolic ketone 1 was then submitted to a Claisen–Schmidt
aldol condensation reaction with each of the three aldehydes 2a–c to
deliver the expected chalcone products 3a–c in moderate to good
yields.11c,14 Thechalcones3a–cwerethencyclizedtogiverisedirectly
to the flavenes 4a–c according to the Clark-Lewis method using so-
dium borohydride in 2-methoxyethanol (Scheme 2).11c,15,17a
The following cis-dihydroxylation of the flavenes 4a–c was
performed using a catalytic amount of osmium tetraoxide in the
presence of N-methylmorpholine N-oxide (NMO)16 to furnish the
expected racemic 2,3-trans-3,4-cis-leucoanthocyanidins 5a–c in
OH
OH OH
anthocyanidins: R3 = H
R1, R2 = H or OH
leucoanthocyanidins
3-GT
anthocyanins: R3 = glucosyl
R1
R1
ANR
LAR
R3 = H
OH
R2
OH
R2
HO
O
HO
O
OH
OH
OH
(2R,3S)-flavan-3-ols
OH
(2R,3R)-flavan-3-ols
Proanthocyanidin Synthase or not ?
R1
OH
HO
O
R2
R1
n
OH
O
4
OH
OH
HO
8
proanthocyanidins
R2
(condensed tannins)
e.g. B type, C4-C8 linkage
OH
OH
Scheme 1. Putative last steps of the biosynthesis of anthocyani(di)ns and
proanthocyanidins.
leucoanthocyanidins that are claimed to be easily converted into
the desired natural isomers under the conditions used for the enzy-
matic assays.5 However, the discrete chemical nature of the mate-
rials thus involved in these assays remains questionable.10b
We thus decided to engage in efforts aimed at the elaboration of
leucoanthocyanidin variants of enhanced stability yet still amenable
to enzymatic studies in vitro. Three flavan-3,4-cis-diol derivatives
(7a–c) convertible into some of the most abundant anthocyanins
found in berries [i.e., malvidin 3-O-glucoside (R1 = R2 = OMe), peoni-
din 3-O-glucoside (R1 = OMe, R2 = H), and pelargonidin 3-O-glucoside
(R1 = R2 = H)]12 were synthesized and grafted onto a biocompatible
solid support (vide infra). The premise behind this approach was that
HO
OH
O
TIPSO
OH
O
TIPSOTf, Et3N
R1
R1
A
CH2Cl2, -78°C, 1h
93%
OR3
R2
OR3
R2
B
B
OH
TIPSO
TIPSO
O
TIPSO
OH
NaBH4
NaH, THF
rt, 2h
a: 63%
b: 82%
c: 43%
1
2
A
C
A
2-methoxyethanol
90 °C, 10 min.
R1
TIPSO
a: R1, R2 = OMe, R3 = nitroveratryl
b: R1 = OMe, R2 = H, R3 = benzyl
c: R1, R2 = H, R3 = benzyl
TIPSO
O
4
OR3
R2
a: 56%
b: 74%
c: 68%
B
3
a: 72%
b: 81%
c: 82%
cat. OsO4, NMO
t-BuOH, THF, rt, 4h
O
OH
R2
R1
R1
R1
4' OR3
R2
OR3
R2
t-Bu2Si(OTf)2,
pyridine
TIPSO
O
TIPSO
O
2
TIPSO
O
a: hν (λ > 300 nm),
CH2Cl2, rt, 3h, 63%
3
CH2Cl2, 0°C to rt, 24h
O
OH
O
or
4
Pd/C, H2, THF, rt
TIPSO
O
t-Bu
a: 77%
b: 81%
c: 93%
TIPSO
OH
TIPSO
O
t-Bu
t-Bu
Si
t-Bu
Si
b: 7h, 86%
c: 24h, 72%
5
7
6
Scheme 2. Synthesis of silylated leucomalvidin (7a), leucopeonidin (7b), and leucopelargonidin (7c).

D. Deffieux et al. / Tetrahedron Letters 50 (2009) 6567–6571
6569
these vicinal hydroxyl groups with triisopropylsilyl group failed,
as only one silyl group could be introduced using TIPSOTf in the
presence of Et3N. The ultimate step consisted in the deprotection
of the 40-hydroxyl group. Thus, the flavan-3,4-diol 6a was sub-
O
O
O
NH2
+
PEGA
i-Pr2NEt, DMF
μW, 50°C, 15 mn
jected to irradiation using
a high-pressure mercury lamp
(k > 300 nm) to furnish the silyl-protected leucomalvidin 7a in
11% overall yield. The corresponding leucopeonidin 7b and leuco-
pelargonidin 7c were obtained after hydrogenolysis of 6b and 6c,
respectively, in 32% and 15% yields over the six steps of their
synthesis.
The instability of free leucoanthocyanidins was here verified by
monitoring via 1H NMR analysis the quasi immediate and steady
degradation of a sample of leucopeonidin 7b in CDCl3 undergoing
desilylation with a solution of TBAF in THF at room temperature
(see Supplementary data).
With adequately protected leucoanthocyanidins 7a–c in hand,
we turned our attention to the selection of a resin on which to graft
them with the following criteria in mind. Such a resin should be
not only appropriately functionalized to enable the attachment of
7a–c via their free phenolic hydroxyl group, but also made of a
polymeric material capable of swelling extensively in aqueous
media for allowing access of proteins (enzymes) into the resin
beads during biochemical assays. Moreover, the loading onto the
resin beads should not be too high to avoid the risk of intermolec-
ular reactions between leucoanthocyanidins once deprotected. The
polyethyleneglycol-polyacrylamide (PEGA) resin with a potential
maximum loading of 0.4 mmol/g meets these criteria.18
O
HO
NH
8
O
7a-c
EDCI, DMAP
DMF, rt,16h
O
R1
O
NH
B
O
R3O
O
2
R2
A
C
3
OR4
4
OR3 OR4
9a-c: R3 = TIPS, R4,R4= -Si(t-Bu)2-
10a-c: R3, R4 = H
Et3N-3HF, THF, rt, 7h
Ac2O, pyridine,rt, 48h
11a-c: R3, R4 = OAc
Scheme 3. Grafting of leucoanthocyanidins onto a functionalized PEGA resin. (a)
R1, R2 = OMe, (b): R1 = OMe, R2 = H; (c): R1, R2 = H.
good yields.17 The 2,3-trans-3,4-cis configuration was confirmed
by 1H NMR analysis, which indicated for each compound a J2,3 cou-
pling constant of about 10 Hz and a J3,4 of 3 to 4 Hz. Next, the cis-
3,4-diol units of 5a–c were protected with a di-tert-butylsilyl
group (Scheme 2). Previous attempts to independently protect
Succinic anhydride was used to install a functionalized linker
compatible with the attachment of 7a–c (Scheme 3). This requisite
adaptation of the PEGA resin was conveniently achieved in the
Figure 1. (Top) 1H NMR spectrum of silylated leucopeonidin 7b in CDCl3; (middle) HR-MAS 1H NMR spectrum of PEGA-supported silylated leucopeonidin 9b in CDCl3; and
(bottom) HR-MAS 1H NMR spectrum of PEGA-supported acetylated leucopeonidin 11b in CDCl3.

6570
D. Deffieux et al. / Tetrahedron Letters 50 (2009) 6567–6571
presence of Hunig’s base in DMF.19 Activation by microwaves per-
mitted to drastically reduce the reaction time from about 5 days to
only 15 min at 50 °C (see Supplementary data). Then, esterification
of the resulting carboxylic acid 8 to the free phenolic function of
7a–c was accomplished in a standard fashion using EDCI/DMAP
as the condensation reagents (Scheme 3). To evaluate the extent
of the loading onto the resin, an aliquot of the PEGA-supported
silylated leucoanthocyanidin 9b was treated with sodium methox-
ide in a CH2Cl2/MeOH (2:1) mixture. The quantity of 7b thus re-
leased and estimated by HPLC analysis indicated a loading onto
the resin of about 50% (i.e., 0.2 mmol/g).
The PEGA-supported silylated leucoanthocyanidins 9a–c were
characterized by high-resolution magic angle spinning (HR-MAS)
NMR analysis. Comparison of their HR-MAS 1H NMR spectra with
the standard 1H NMR spectra of the silylated leucoanthocyanidins
7a–c in solution indicated an excellent level of concordance be-
tween the diagnostic signals of these pairs of spectra (see Fig. 1
and Supplementary data). The absence of methoxy signal(s) in
the spectra of 9a and 9b (see Fig. 1 for 9b, middle spectrum) is a
consequence of the presaturation applied at 3.6 ppm to suppress
the overwhelming ethylene signals from the PEG chains of the
resin.20
leucoanthocyanidins 10a–c are thus granted with a much higher
stability than their free counterparts, since they survive successive
exposures to desilylation, HR-MAS NMR analysis, and acetylation
conditions (see Supplementary data).22
In summary, the synthesis of three leucoanthocyanidin deriva-
tives immobilized on a biocompatible PEGA resin via their 40-hy-
droxyl group was achieved. These materials constitute stabilized
variants of these otherwise fleeting flavan-3,4-diol species that
should serve, in spite of their racemic nature, as valuable tools
for in vitro enzymatic studies aimed at elucidating the final steps
of the biosynthesis of both anthocyanidins and proanthocyanidins.
Affinity and biochemical reactivity tests with LDOX are in progress
and the results will be reported in due course.
Acknowledgments
We wish to thank the Conseil Interprofessionnel du Vin de Bor-
deaux for their generous financial support, including research
assistantships for Sophie Gaudrel-Grosay and Céline Chalumeau.
This project was undertaken within the framework of the associa-
tion of our research team with the Institut des Sciences de la Vigne
et du Vin, Bordeaux-Aquitaine, France.
The desilylation of 9a–c caused us some difficulties, as the use
of TBAF in THF or cesium fluoride in DMF consistently led to cleav-
age of the ester linkage to the resin.21 We thus had to rely on the
use of Et3N–3HF in THF at room temperature (Scheme 3), but the
corresponding HR-MAS 1H NMR spectra run in either CDCl3 or a
CDCl3/DMSO (1:1) solvent mixture exhibited such a low resolution
that confirmation of the efficiency of this desilylation step leading
to 10a–c could not be made by this means. Nevertheless, good
quality spectra were obtained in CDCl3 after peracetylation leading
to the PEGA-supported leucoanthocyanidin derivatives 11a–c (see
Fig. 1 for 11b, bottom spectrum). The coupled proton spin system
of their cycle C (i.e., H2, H3, and H4) could clearly be assigned by 2D
COSY HR-MAS NMR experiments (see Fig. 2 for 11b). These satis-
factory HR-MAS NMR analyses of 11a–c not only confirmed the
success of the preceding desilylation step (Scheme 3), but also
provided a sound indication that the resulting PEGA-supported
Supplementary data
Supplementary data (experimental procedures, characteriza-
tion data, and NMR spectra for all major compounds) associated
with this article can be found, in the online version, at
doi:10.1016/j.tetlet.2009.09.045.
References and notes
1. (a) Forkmann, G.; Heller, W.. In Comprehensive Natural Products Chemistry;
Sankawa, U., Ed.; Pergamon: Oxford, 1999; Vol. 1, pp 713–748; (b) Davies, K.
M.; Schwinn, K. E. In Flavonoids: Chemistry, Biochemistry and Applications;
Andersen, Ø. M., Markham, K. R., Eds.; CRC-Taylor & Francis: Boca Raton, FL,
2006; pp 143–218.
2. Holton, T. A.; Cornish, E. C. The Plant Cell 1995, 7, 1071–1083.
3. (a) Xie, D.-Y.; Sharma, S. B.; Paiva, N. L.; Ferreira, D.; Dixon, R. A. Science 2003,
299, 396–399; (b) Xie, D.-Y.; Dixon, R. A. Phytochemistry 2005, 66, 2127–2144.
4. (a) Williams, C. A.; Grayer, R. J. Nat. Prod. Rep. 2004, 21, 539–573; (b) Ferreira,
D.; Nel, R. J. J.; Bekker, R.. In Comprehensive Natural Products Chemistry; Pinto, B.
M., Ed.; Pergamon: Oxford, 1999; Vol. 3, pp 747–797.
5. (a) Saito, K.; Kobayashi, M.; Gong, Z.; Tanaka, Y.; Yamazaki, M. Plant J. 1999, 17,
181–189; (b) Turnbull, J. J.; Sobey, W. J.; Aplin, R. T.; Hassan, A.; Firmin, J. L.;
Schofield, C. J.; Prescott, A. G. Chem. Commun. 2000, 2473–2474; (c) Nakajima,
J.; Tanaka, Y.; Yamazaki, M.; Saito, K. J. Biol. Chem. 2001, 276, 25797–25803.
6. (a) Wellmann, F.; Griesser, M.; Schwab, W.; Martens, S.; Eisenreich, W.; Matern,
U.; Lukacin, R. FEBS Lett. 2006, 580, 1642–1648; (b) Turnbull, J. J.; Nakajima, J.;
Welford, R. W. D.; Yamazaki, M.; Saito, K.; Schofield, C. J. J. Biol. Chem. 2004, 279,
1206–1216.
7. Dixon, R. A.; Xie, D.-Y.; Sharma, S. B. New Phytol. 2005, 165, 9–28.
8. Tanner, G. J.; Francki, K. T.; Abrahams, S.; Watson, J. M.; Larkin, P. J.; Ashton, A.
R. J. Biol. Chem. 2003, 278, 31647–31656.
9. (a) Tian, L.; Pang, Y.; Dixon, R. A. Phytochem. Rev. 2008, 7, 445–465; (b) He, F.;
Pan, Q.-H.; Shi, Y.; Chang-Qing, D. Molecules 2008, 13, 2674–2703.
10. (a) Halbwirth, H.; Kahl, S.; Jäger, W.; Reznicek, G.; Forkmann, G.; Stich, K. Plant
Sci. 2006, 170, 587–595; (b) Stafford, H. A.; Lester, H. H.; Porter, L. J.
Phytochemistry 1985, 24, 333–338.
11. (a) Porter, L. J.; Foo, L. Y. Phytochemistry 1982, 21, 2947–2952; (b) Steenkamp, J.
A.; Ferreira, D.; Roux, D. G. Tetrahedron Lett. 1985, 26, 3045–3048; (c) Marais, J.
P. J.; Ferreira, D.; Slade, D. Phytochemistry 2005, 66, 2145–2176. and references
cited therein.
12. (a) Andersen, Ø.; Jordheim, M. In Flavonoids: Chemistry, Biochemistry and
Applications; Andersen, Ø. M., Markham, K. R., Eds.; CRC-Taylor & Francis: Boca
Raton, FL, 2006; pp 471–551; (b) Munõz, S.; Mestres, M.; Busto, O.; Guasch, J.
Anal. Chim. Acta 2008, 628, 104–110.
13. Patchornik, A.; Kraus, M. A. J. Am. Chem. Soc. 1970, 92, 7587–7589.
14. (a) Claisen, L.; Claparède, A. Ber. Dtsch. Chem. Ges. 1881, 14, 2460–2468; (b)
Schmidt, J. G. Ber. Dtsch. Chem. Ges. 1881, 14, 1459–1461.
15. (a) Clark-Lewis, J. W.; Jemison, R. W.; Skingle, D. C.; Williams, L. R. Chem. Ind.
1967, 1455–1456; (b) Clark-Lewis, J. W.; Skingle, D. C. Aust. J. Chem. 1967, 20,
2169–2190.
16. For a review on osmium tetraoxide-mediated cis-hydroxylation of olefins, see:
Schröder, M. Chem. Rev. 1980, 80, 187–213.
Figure 2. Diagnostic portion of 1H–1H HR-MAS COSY map of PEGA-supported
acetylated leucopeonidin 11b in CDCl3.

D. Deffieux et al. / Tetrahedron Letters 50 (2009) 6567–6571
6571
17. (a) Kawamoto, H.; Nakatsubo, F.; Murakami, K. J. Wood Chem. Technol. 1989, 9,
35–52; (b) Nay, B.; Arnaudinaud, V.; Peyrat, J.-F.; Nuhrich, A.; Deffieux, G.;
Mérillon, J.-M.; Vercauteren, J. Eur. J. Org. Chem. 2000, 1279–1283.
18. (a) Meldal, M. Tetrahedron Lett. 1992, 33, 3077–3080; (b) Auzanneau, F.-I.;
Meldal, M.; Bock, K. J. Pept. Sci. 1995, 1, 31–44.
20. Keifer, P. A. J. Org. Chem. 1996, 61, 1558–1559.
21. Curtis, A. D. M.; Humphries, M. J.; Ramsden, C. A. ARKIVOC 2000, iii, 218–227.
22. However, for long-term storage, these PEGA-supported leucoanthocyanidins
were kept under their silylated forms 9a–c in the freezer at ꢀ18 °C. After
several weeks of storage under these conditions, HR-MAS 1H NMR analysis also
indicated no degradation of these materials.
19. Fruchart, J.-S.; Grass-Masse, H.; Melnyk, O. Tetrahedron Lett. 1999, 40, 6225–
6228.